thadec

Marvin said:

The high performance computing pro market is a rounding error in the computer industry and has been for a long time.

Not even close to being true. The workstation, small server and HEDT markets are booming. There was actually a Threadripper shortage last year. Intel launched Sapphire Rapids (Xeon W-2400 and W-3400) this year to compete with Threadripper in the desktop workstation market. AMD is going to launch the Threadripper 7900X this year along with the usual Pro 7900WX in order to compete with Intel in HEDTs. I am still wondering what is keeping Nvidia from getting in on the act with a stripped down version of their Grace chips targeting this market.

Engineers, architects, researchers etc. have always needed workstations. Do you think that these entire professions stopped existing? Moved to the cloud? Or can get their work done on MacBooks now? And now they have company. Thanks to YouTube and all that, the number of people into serious video editing and computer animation has gone through the roof. Add to those the AI/ML boom the past few years and now the LLM types is going to mean still more. Yet rounding error you say. 

 Apple not being able to do the Extreme until TSMC's 2nd gen 3nm process is ready until 2024, which forced them to just stick an M2 Ultra in a cheese grater and call it a day because needing to move forward with Sonoma meant that they couldn't wait any longer, is no reason to just go and make up stuff OK?

AppleInsider said:

Apple Silicon has been great for Apple, and the Mac faithful. It's got industry-leading computing grunt to power consumption, and it does it all quietly and in a very cool package. No more lap-singeing MacBook Pros!

It's restored pricing at the low-end that hasn't been seen in a long time. It's also broken the shackles of promises Intel made for years that led to engineering choices that couldn't handle the heat as well as they should have been able to when the chips that fell short of those lofty goals were delivered.

Honestly guys this isn't true. At all. The consensus - see Tom's Hardware and Wccftech is that the M2 Ultra's Geekbench scores are essentially the same as the Intel Core i9-13900K (Geekbench 5 has the i9 slightly higher, Geekbench 6 the Ultra slightly higher). Now the Core i9 is a chip primarily for desktop gamers. A hypothetical Mac Pro refresh would have had a Xeon W9-3495X that crushes the Core i9 and should have had an AMD Threadripper 7000 that when it launches in September crushes the Xeon W9. Claiming that an Intel Core i9 system with 256 GB RAM, 2 TB SSD and an Nvidia RTX A5000 (or AMD Radeon Pro W6800X) will set you back $10,000 just isn't being honest. 
Apple Silicon is only a better value than AMD systems in particular when you don't have the added expense of a discrete GPU, like the $599 Mac Mini whose integrated GPU is better than the entry level AMD, Nvidia and Intel discrete options. Otherwise x86 systems perform better and are cheaper.

When the M3 Extreme arrives next year, maybe. But an M2 Ultra workstation with CPU performance no better than a desktop PC chip and graphics performance akin to a midlevel pro dGPU? Not a chance.

JamesCude said:

The Mac Pro becomes even more of a niche product with the Mac Studio out there. The Studio offers more than enough power for most use cases and the need for PCI cards is rarer than ever. 

Too bad- it’s awesome to have gobs of power but with the insane efficiency of Apple Silicon it’s no longer necessary. You can have your cake and eat it already.

Sorry, but this isn't even close to being true. The workstation and small server markets were already large before the boom in data science, ML and AI, and now with the LLM stuff it is going to get even bigger. Then there are the traditional use cases mentioned by @mikethemartian - physics/engineering, medical research, financial/economics modeling applications - to which we can also add CAD, architecture, animation, VFX etc. A lot of those require the most RAM and best GPUs that you can throw at them (for parallel processing, not actual graphics). In some cases integrated approach makes things worse. Discrete GPUs come with their own memory. While Apple's integrated CPU, GPU and RAM approach makes certain things faster and reduces power consumption, it creates bottlenecks when both the CPU and GPU are trying to access the same memory pool. And that is with 192 GB RAM not being nearly enough anyway. 

That being said, I have left several comments on here to this effect: a Mac Pro is whatever Apple calls a Mac Pro. The first iPhone had a 3.5" screen, 128 MB RAM, a single core RISC SOC and the original plan was no app store with an emphasis on HTML5 apps. Look at an iPhone 14 Pro Max by comparison. Also, compare the original iPod to the iPod Touch. The original Mac Mini to the M2 Pro Mac Mini. The original Apple TV to the current one. The iMac G3 to the current iMac. And so on. 

Apple can openly concede that the old Xeon W-based Mac Pro was a failure - too big, too noisy, too expensive, had crazy power/cooling requirements, was rarely updated, sold in much lower volumes than workstations from HP, Lenovo, Dell etc. - and the concept is being scrapped. And they can market the new Mac Pro as a innovative computing segment with use cases and markets that they define. 

Why do this? Because frankly ... they don't have a choice. First, Intel has made up a lot of ground in a very short time. Second, the best workstations no longer have Intel Xeon W inside anyway. They have AMD Threadripper. Would even an M3 Extreme Mac Pro outperform the AMD Threadripper 7000 that we are going to get in September? If so, it will only be due to the 7000 still being on a 5nm process. Meaning that when the 3nm Threadripper 8000 comes out in early 2025? Apple won't be able to compete with general purpose Intel and AMD workstations on CPU, GPU or RAM. So, position it as a special purpose device that is better for the things that Apple claims in the ways that Apple says that it is. 

Good grief. There is so much wrong here.

Unlike the x86 chip architecture from Intel, designed to be general-purpose processors, ARM-based system-on-a-chips (SoC) are highly customizable. Companies can design these processors as Apple has to include more high-performance CPU cores and integrated memory to compete with x86 CPUs.

Totally ignoring that Apple went to Intel to design the Core Duo for the MacBook Air. Or that Microsoft and Sony get custom-made chips from AMD for their XBox and PlayStation consoles every generation. Or that Valve and Asus went to AMD to get custom-made chips for their consoles. 

The custom cores allow advanced features such as artificial intelligence and machine learning tasks, such as Apple's Neural Engine it includes in its processors.

Yeah ... Intel and AMD have their own integrated NPUs also.

ARM chips can also be more efficient for power and customized to meet specific power requirements.

As can x86 chips. The latest AMD 7040 laptop chips give better power per watt than the M1 and M1 Pro. Intel has a whole range of laptop chips that give better power per watt than the iPad Pro does.

Integrating GPUs into the chips also improves performance by accelerating computational tasks such as machine learning and image recognition. As a result, it allows for faster and more accurate results in computing and makes it possible to run advanced apps and software on ARM-based computers.

Literally every single Intel and AMD laptop chip has an integrated GPU and has for decades. Most Intel desktop chips have integrated GPUs too. Only AMD desktop chips regularly lack integrated GPUs. Are they as good as Apple's integrated GPUs? Not yet, though the integrated GPU on the latest AMD laptop chips perform as well as an Nvidia GTX 1650: https://wccftech.com/amd-radeon-780m-rdna-3-igpu-comes-within-striking-distance-of-nvidia-gtx-1650-dgpu/ But they don't have to. Because - unlike Apple Silicon - you have the option of using discrete GPUs with Intel and AMD chips. 

As for the unified memory thing: face reality. First off, no one needs it. Apple uses unified memory to maximize performance. Intel and AMD can get the performance they need without it. Second, no one wants it. People prefer the ability to cheaply expand RAM on low and midrange systems. And they prefer the ability to add massive amounts of RAM on high end systems and servers. Being able to buy an entry level system for cheap and upgrade the RAM to 64 GB for $100 is huge. And this site states that actual machine learning workstations often require 1 TB of RAM https://www.pugetsystems.com/solutions/scientific-computing-workstations/machine-learning-ai/hardware-recommendations/#ram so you can forget about the oroposed 192 GB RAM Mac Pro being used by that crowd. 

Look, I am not saying that the people who write for Apple Insider need a computer engineering background or anything like that, but so much in this article can be easily debunked just by spending 30 minutes on Anandtech. At least the magical and wishful thinking that Daniel Eran Dilger was infamous for on Apple Insider was mostly opinion.

Yeah ... Intel and AMD CPUs have supported both integrated graphics (AMD's RDNA 3 integrated graphics on their Ryzen 7040 laptop chips are comparable to an Nvidia RTX 2050) as well as discrete graphics through PCIE or Thunderbolt for who knows how long (and now M.2 slots). Intel drivers will even have an Intel Arc discrete GPU and an Intel Iris Xe integrated GPU be seen by the CPU as a single GPU. (AMD considered this idea but abandoned it.) And no, it isn't an x86 thing. Lots of Linux ARM servers use discrete GPUs. MediaTek and Nvidia wanted to bring discrete GPU support to ARM PCs around 2021 but abandoned it because neither Microsoft (who has an exclusive Windows on ARM deal with Qualcomm that is explicitly designed to lock out MediaTek for Qualcomm's sake and ChromeOS/Linux for Microsoft's sake) or Google (who just isn't very smart when it comes to stuff like this) were interested so it was abandoned.

So, there never has been any reason for Apple Silicon Macs not supporting discrete graphics via M.2, PCIE or Thunderbolt other than Apple simply not wanting to. Which was the same reason why Apple locked Nvidia out of the Mac ecosystem and had people stuck with AMD GPU options only: purely because they wanted to. My guess is that Apple believed that they were capable of creating integrated GPUs that were comparable with Nvidia Ampere and AMD Radeon Pro, especially in the workloads that most Mac Pro buyers use them for. Maybe they are, but the issue may be that it isn't cost-effective to do so for a Mac Pro line that will sell less than a million units a year.